Tutorial on the Impact of the Synchronous

Generator Model on Protection Studies

Normann Fischer, Gabriel Benmouyal, and Satish Samineni

Schweitzer Engineering Laboratories, Inc.

Published in
SEL Journal of Reliable Power, Volume 3, Number 1, March 2012

Previously presented at
Waterpower XVI, July 2009

Previous revised edition released June 2009

Originally presented at the

35th Annual Western Protective Relay Conference, October 2008
 1

Tutorial on the Impact of the Synchronous

Generator Model on Protection Studies
Normann Fischer, Gabriel Benmouyal, and Satish Samineni, Schweitzer Engineering Laboratories, Inc.

Abstract—The classical theory of representation of power In steady state, the current flowing in the rotor winding is
swings in the impedance plane is based on the representation of equal to the exciter voltage divided by the winding resistance:
synchronous generators as constant voltage sources. The classical
model of a synchronous generator represents the machine as a Vf
If = (2)
constant voltage source behind the transient reactance in the rf
direct axis. The classical model of synchronous generators is
based on the assumption that the rotor flux linkage will not The field winding has a self-inductance Lff. A fundamental
change during a short period of time following a major characteristic of a synchronous generator is the direct-axis
disturbance. In reality, with constant excitation voltage, the rotor open-circuit transient time constant T'd0, the ratio of the field
flux linkage will decrease and the internal generator voltage will self inductance over its dc resistance:
decrease accordingly. The addition of an automatic voltage
regulator (AVR) boosts the excitation voltage following a Lff
'
Td0 = (3)
disturbance so that the rotor flux linkage will be sustained and rf
the generator internal voltage will be prevented from collapsing.
The purpose of this paper is to show how power swing The order of magnitude of this time constant, which is
representation in the impedance plane will depart from the typically a few seconds, indicates that the voltage at the
classical theory when complex AVRs are used on modern synchronous generator terminals cannot be changed
generators. instantaneously; in other words, the current in the field
winding varies according to the field open-circuit time
I. BASIC SYNCHRONOUS GENERATOR PHYSICAL PRINCIPLES constant.
This paper is not intended to be an exhaustive review of When the generator is loaded, the three-phase currents will
synchronous generator physical and engineering principles, create a flux, represented in Fig. 2 as ψia+ib+ic. This flux is
but rather an overview of fundamental and essential facts. It is known as the armature reaction. The vectorial addition of the
assumed here that the reader is familiar with the two-axis field flux and the armature reaction is the air-gap flux, which
model representation of synchronous machines [1]. is represented as ψag.
Direct Quadrature
Axis ia Axis Direct Quadrature
Axis ia Axis
ψf
ψf ψ
ag
–ic Eq –ib
–ic –ib
If
Stator If
Phase ψ
Vf Windings ia + ib + ic
vf

ib ic
ib ic

–ia
–ia
Fig. 1. Synchronous generator basic mechanical and electrical structure
Fig. 2. Armature reaction
Fig. 1 represents a two-pole synchronous generator. The
The projection of the air-gap flux on the direct and
rotor winding is supplied with a dc exciter voltage Vf. A dc
quadrature axis, as in Fig. 3, defines the fluxes in the direct
current If flows in the rotor winding to create a flux ψf in the
and quadrature axes as ψd and ψq.
generator direct axis. When the primary mover drives the
The fluxes in the direct and quadrature axes create the
generator and the generator rotates at synchronous speed, the
corresponding voltages vd and va along the same axes. We can
flux will induce a voltage in the three-phase stator windings.
demonstrate [1] these voltages with:
When the generator is unloaded, we can measure an excitation
or generator internal voltage at the terminals. This internal dψ d
vd = ra i d + − ωψ q (4)
voltage, Eq in Fig. 1, is proportional to the current If and lies in dt
the quadrature axis 90 degrees ψf. Eq is given by (1). and
E q = ω M f If (1)
 2

dψ q or, if it changes, it will change slowly because of the long time

vq = ra i q + + ωψ d (5) constant associated with the rotor.
dt
Consider the exciter voltage we obtain from (9):
Direct Quadrature dψ f
Axis ia Axis E′ = vf = R f if + (9)
ψf
dt
ψ
ag
A new exciter voltage referred to the armature is defined in
–ic –ib (10):
ψ
q ω M f E ′ex
If
ψ
d E ex = (10)
Rf
vf
This new exciter voltage will have a value of 1 pu when the
generator is open circuited and its terminal voltage is 1 pu. If
ib ic
we assume that the field flux does not vary following a
disturbance, we can define a new fictitious voltage
–ia proportional to ψf that also does not vary in (11) [1]:
ω Mf
Fig. 3. Fluxes in the direct and quadrature axis E ′q = ψf (11)
Lff
A. The Synchronous Generator in the Steady State
In steady state, we can express the voltage phasor at the We can then relate the new fictitious voltage to the
generator terminals in terms of the phasors in the direct and excitation voltage with (12) [1]:
quadrature axes as follows: E ′q = E q − (x d − x ′d )i d (12)
Vt = Vd + jVq (6) Fig. 5 shows a salient pole generator vector diagram in
In the same fashion, we can express the current phasor after transient state by indicating the relation between Eq and E'q.
it has been projected on the two axes as follows: In the transient state and following a disturbance, if we
assume that the internal voltage E'q remains constant, we can
I t = Id + jIq (7) show the next relation between the internal voltage E'q and the
In steady state, (8) provides the vectors’ relation linking the generator terminal voltage Vt with (13):
excitation voltage to the terminal voltage and current for a E ′q = Vt + jX ′d Id + jX ′q Iq (13)
salient-pole machine.
If we assume that the saliency is removed, in other words,
E q = Vt + ra I t + j x d Id + j x q Iq (8) X'd = X'q, we can define a new constant voltage internal E'i
Fig. 4 shows the corresponding generator vector diagram. that is equal to (14):
In this figure, δ is the internal angle between the excitation E i′ = Vt + jX ′d I t (14)
voltage and the generator terminal voltage. In the same figure,
This last voltage can represent a generator as a constant
φ is the angle between the generator terminal voltage and voltage source behind transient reactance, as in Fig. 6. We can
current that determines the power factor. therefore represent a generator as a constant voltage source
behind the transient impedance if the following assumptions
Iq Eq are in effect:
• The rotor flux linkage remains constant.
δ • Saliency is removed. In other words, there exists only
j xq Iq
one reactance that is X′d.
φ Vt
Id
It j (Xd – X'd) Id
Ra It j xd Id

Iq j X'd It E'q = E' Eq

Fig. 4. Salient-pole generator vector diagram in steady state

E'i
δ
B. The Synchronous Generator in the Transient State
Vt j Xq Iq = j X'q Iq
Based on the relations between the different vectors in the
Id
steady state as in Fig. 4, the fundamental theorem of the flux It

linkage provides some insight about the behavior of a j X'd Id

generator in transient state. The theorem of flux linkage states Fig. 5. Salient-pole generator vector diagram in transient state
that “the flux linkage of any closed circuit of finite resistance
and finite e.m.f. cannot change instantly” [1]. In reality, the
flux linkage will vary according to the circuit time constants.
Applying this principle to a synchronous generator, one could
state that the field flux does not change during a disturbance
 3

X'd
3.4

E'i Vt E = 3.5
3.2 ex

Fig. 6. Representation of a synchronous generator as a constant voltage 3

source behind transient reactance E = 2.0
ex
If we assume that E′q is not constant, a relation exists that 2.8
provides the rate of change of this fictitious voltage with (15) E = 1.0
ex
[1]: 2.6

dE ′q E ex − E q 0 0.1 0.2 0.3 0.4 0.5

= (15)
dt ′
Td0 Time (s)

When we do not assume a constant flux linkage following Fig. 7. Eq variation with different excitation voltages Eex
a disturbance, the rate of change of the voltage proportional to
1.2
the rotor flux linkage is as in (15): a function of the exciter
voltage Eex, the excitation voltage Eq, and the generator rotor
1.1
open circuit transient time constant T′d0. Note that as the rotor
time constant T′d0 increases, the rate of change of E′q E = 3.5
ex
decreases. 1
E = 2.0
To illustrate the impact of exciter voltage Eex on Eq and E′q ex
following a disturbance close to the generator, consider the 0.9
example of an unloaded generator, with unit terminal voltage, E = 1.0
subjected to a three-phase short circuit at its terminals. The ex
0.8
generator impedances are as in (16):
X d = 1.15, X ′d = 0.37, X q = X ′q = 0.75 (16) 0.7
0 0.1 0.2 0.3 0.4 0.5
Because the generator is unloaded, the identities in (17) are Time (s)
applicable prior to the short circuit:
Fig. 8. E′q variation with different excitation voltages Eex
Vt = E ex = E q = E ′q = 1.0
(17) C. The Notions of Synchronizing and Damping Torques
It = 0
Consider the simple network model of Fig. 9, where a
If we follow the methodology indicated in [1], Fig. 7 shows generator represented as a constant voltage source behind the
the variation in time of the excitation voltage Eq when the transient reactance supplies an infinite bus through an
short circuit is applied with different exciter voltages. At the impedance Xe. This generator model is the classical model of
moment the short circuit is applied, Eq will jump from 1.0 to a generator.
3.29 because of the fault current. Eq will then vary according
to the exciter voltage applied. In manual mode (Eex = 1.0), the X′d Vt Xe
excitation voltage will start dropping linearly. With a voltage
regulator, as the exciter voltage jumps to its ceiling value (see
Section II), the rate of change of the internal voltage drop XT
decreases. The figure shows the variation of Eq with two E i′ ∠ δ Es ∠ 0
regulator ceiling voltages: 2 and 3.5 pu. There exists a value
for the exciter voltage (slightly smaller than 3.5 pu for the
example) where the excitation system will sustain Eq. Above
this value, Eq will increase linearly during the short circuit.
Fig. 8 uses the same example during the three-phase short Fig. 9. Elementary power network with classical representation of generator
circuit to illustrate the variation of E′q. Following the We define the following:
application of the short circuit, E′q does not change, because of
∆Tm is the variation of the mechanical power input
the constant flux linkage principle. As for Eq, E′q will drop to the generator in per unit (pu).
linearly with low values of the exciter voltage Eex. For Eq,
there is a threshold value for Eex where E′q will remain H is the inertia constant in seconds.
unchanged during the short circuit. Above the threshold, E′q M is the inertia coefficient = 2H in seconds.
will start increasing linearly rather than decreasing.
This example illustrates the impact of an autoregulator on ω0
is the base rotor electrical speed in radians per
generator internal voltage during a disturbance. An auto- second (377 rad/s).
regulator can increase the generator internal voltages during a The control block diagram that Fig. 10 represents allows us
disturbance. This capacity is directly related to improving the to use the technique of small-signal analysis [2] [5] to study
generator transient stability, as this paper explains later. the dynamics of the elementary network in Fig. 9.

The total electrical torque that opposes the mechanical
power input the synchronous machine produces is the sum of
two synchronous and damping torques and is equal to (18):
∆Te = ∆Tsync + ∆Tdamp = K1 ∆δ + K D ∆ω
The synchronous torque is proportional to the machine
internal angle variation in (19):
∆Tsync = K1 ∆δ
The damping torque is proportional to the machine speed
variation in (20):
∆Tdamp = K D ∆ω
For the system in Fig. 9, we can demonstrate that K1 is
equal to (21) [2]:
 E′ E 
K1 =  i s  cos δ0 (21)
 XT 
For the generator belonging to the network in Fig. 9 to be
stable, both the synchronous and damping torques have to be
positive. A lack of either of the two torques will render the
generator unstable.
K1
∆Tsync = synchronizing
torque
1 ∆ω 377
∆Tm (pu)
Ms s
∆Tdamp = damping
torque KD
Fig. 10. Control block diagram of an elementary power network
II. SYNCHRONOUS GENERATOR EXCITATION SYSTEMS
A. The Manual and Automatic Voltage Regulator (AVR)
The primary function of a synchronous generator excitation
system is to regulate the voltage at the generator output. This
paper focuses on present-day excitation static systems such as
the one in Fig. 11.
In these systems, the input power for the static exciter is
commonly derived from the machine terminals. A step-down
transformer (excitation transformer PPT) feeds a three-phase
controlled rectifier bridge that converts ac voltage into dc
voltage. The dc output is connected to the machine field
winding by brushes and collector rings.
In manual mode, either the level of the generator output
voltage or the field current level (as in Fig. 11) is under the
manual control of the operator. Although manual control of
the excitation system still occurs on some old machines,
organizations such as the North American Electric Reliability
Corporation (NERC) recommend against this practice today
because of the drawbacks and shortcomings that this mode of
operation entails.
4
In automatic mode, a voltage set point is introduced in the
summing point of the AVR. The excitation system compares
this voltage set point to the generator output voltage
(18) measurement, and the comparison produces an error signal
that adjusts the timing of the firing of the silicon-controlled
rectifiers until the output voltage Vt equals the voltage set
point. In steady state, the generator output voltage is therefore
(19) equal to the voltage set point.
SCR Bridge
Field PPT
(20) Circuit Terminal
Voltage Vt
PT
– +
Manual
Regulator
Set Point If
– +
Automatic
Regulator
Set Point Et
Fig. 11. AVR principle with self-exciting generator
Fig. 12 represents the generic model of a static excitation
as provided among others by [10]. Such models are intended
as guidelines for stability studies. Vref is the voltage setting,
∆δ and VC is the voltage measurement from the generator
terminals. The difference between these two quantities
constitutes the basic error signal. These models provide for
additional error signals at the AVR summing point. VS is the
error signal from a power system stabilizer (PSS). VUEL is the
error signal from an under excitation limiter, which we
describe later. In the excitation system of Fig. 12, an auction
occurs between some signals; in other words, a high-voltage
(HV) gate will pick out the input signal that has the highest
level when a low-voltage (LV) gate picks out the signal that
has the smaller one. This auctioneering action, when it occurs,
allows some signals to take control of the AVR loop. As an
example, following the AVR summing point, if the error
signal from the UEL circuit is larger than the error signal from
the summing point, the system gives priority to the UEL
signal that takes control of the AVR loop. The output of the
AVR is the voltage supplied to the field circuit. This voltage is
bound and is of primary importance. The maximum voltage
the excitation system supplies is commonly called the AVR
ceiling. During a severe disturbance, such as a short circuit
close to the generator terminals, the system will most likely
apply this exciter ceiling voltage as the field voltage to
counteract decreasing generator output voltage. Section I
provides an example to demonstrate how the voltage regulator
can sustain the generator internal voltage during a short circuit
at the generator terminals.
 5

VUEL V*UEL V*UEL VOEL

VAMAX VS*
VS (Et VRMAX - KC IFD)
VIMAX
VA
VI HV (1 + s TC )(1 + sTC1 ) KA HV LV
Vref + + EFD
Gate (1 + s TB )(1 + sTB1 ) 1 + s TA
–
Gate Gate
– –
VC
VIMIN VAMIN Et VRMIN

VF KF
1 + s TF

KLR + IFD
V*UEL, V*OEL: Alternate Inputs –
0
ILR

Fig. 12. IEEE type ST1A excitation system

Fig. 13 represents a simplified model of an AVR by a gain C. The Underexcitation Limiter

KA, again, with a time constant.
Vref KA
+ EFD
– 1 + sTA
Terminal Voltage, Vc
Fig. 13. Simplified representation of a static excitation system
B. The Power System Stabilizer
One negative effect of an AVR installed on a generator is
that it decreases the internal damping torque when its gain KA
is increased (the synchronizing torque is, however, increased).
Because of this, the regulator gain must be limited to some
value between 15 and 25 in most situations [4], and this
limitation then reduces the dynamic stability of the generator.
Two solutions exist for this AVR gain limitation: 1) limit the
AVR gain or 2) supplement the AVR with a PSS.
The PSS basic principle, as Fig. 15 shows, consists of
measuring the change with respect to synchronous speed and
sending a signal derived from this speed variation to the
summing point of the AVR [2]. The net effect of the PSS
action is to increase the generator damping torque in both
steady and transient state. Another result of using the PSS is
being able to increase the AVR gain KA without affecting the
overall generator stability.
Vref
+
The purpose of an underexcitation limiter (UEL) is to
prevent a generator from being operated in steady state in a
determined underexcited region.
Consider the UEL model (type UEL2) in Fig. 15. This
model comes from the recommended models in [11]. We can
determine the UEL static or steady-state characteristic by
setting the Laplacian operator “s” to zero and by looking at the
condition when the error signal from the UEL circuit will be
zero [1]. Equation (22) provides this condition:
P E −t k1 KUP − E kt 2 KUV − Q E −t k1 KUQ = 0 (22)
Expressing Q as a function of P, we obtain (23):
KUP + k 2) KUV
Q=P − E (k1
t (23)
KUQ KUQ
Equation (23) describes a straight line, as shown in Fig. 16,
and represents the UEL characteristic in the P-Q plane. When
the generator operating point falls below the line segment, the
UEL produces a positive error that the system supplies to the
AVR summing point. This positive error, in turn, has the
effect of increasing the voltage setting or AVR voltage
reference so that the generator terminal voltage increases until
the generator operating point goes above the UEL limit
straight-line characteristic [3] [5].
KUQ
QxF1
Q
1 + sTUQ

Vfmax
F2 KUV VUEL to AVR
Terminal
KA F2 = (Et)k2 Summing
Vf 1 + sTUV
Voltage Vc – 1 + sTA et F2 = (Et)–k2 Point
+

Vfmin
PxF1 KUV
Phase P
1 + sTUV
Gain Washout Compensation
Fig. 15. Example of a type UEL2 straight line UEL
sTw 1 + sT1
∆ω R KPSS
1 + sTw 1 + sT2

Fig. 14. Principle of a PSS

 6

Q
Xd, Xq Xe
KUV
KUP
P
Eq Vt Es
KUV (k1+k2) Fig. 17. Elementary generator system
Vt
KUQ KUP
slope =
KUQ

j (Xd–X′d) Id
Iq j x′d It E′q=E′ Eq
Fig. 16. Type UEL2 straight line characteristic

Reference [11] describes two additional UEL δ E′i

characteristics, one circular (type UEL1) and one Vt j xq Iq=j x′q Iq

multisegment straight line (type UEL3), that work on the same
principles as type UEL2.
Keep in mind that, while the AVR and the PSS will
improve the dynamic stability of a generator, the purpose of a
UEL is to help prevent the generator from undergoing steady-
state instability resulting from its operation in the
underexcited region.
D. High-Speed, High-Ceiling Voltage Excitation Systems
A network disturbance resulting from a fault close to a
generator reduces the generator’s terminal voltage. If the
generator excitation system is operated in manual mode, its
internal voltage will decrease according to (15).
If the generator excitation system is under the control of an
AVR, the voltage the AVR imposes on the field winding
during the fault will depend upon the AVR speed, gain, and
ceiling voltage. Consequently, the amount of boost that the
generator internal voltage will receive depends upon these
three factors.
It is generally recognized that a high-speed, high-gain, and
high-ceiling AVR supplemented with a PSS is presently one
of the best means to improve generator transient stability [2].
III. REVIEW OF CLASSICAL STEADY-STATE
AND TRANSIENT STABILITY METHODS
A. Steady-State Stability
We can define the steady-state stability limit (SSSL) of a
particular circuit of a power system as the maximum power at
the receiving end of the circuit that we can transmit without
loss of synchronism if we increase the load in very small steps
and if we change the field currents after each increment so as
to restore normal operating conditions [1].
Consider the elementary system of Fig. 17, which consists
of a generator with constant internal voltage Eq that supplies
an infinite bus through an impedance Xe. The conventional
formula in (24) provides the steady-state power transfer
equation for a salient-pole machine:
Eq Es Xd − Xq
P= sin δ + E s2
Xd + Xe 2 (X d + X e ) (X q + X e )
In (24), δ is the angle between Eq and Es, as in Fig. 18.
Id
It
Es j x′d Id
j Xe It
Fig. 18. Vector diagram of network in Fig. 17
Equation (24) allows us to determine the maximum power
that we can transfer from the generator before reaching
steady-state instability. We can also plot the stability limit in
the P-Q plane. With manual operation, and if we assume that
saliency has been removed, the classical SSSL is a circle with
center and radius as in Fig. 19 [3] [5].
Q (pu)
Vt2 1 1
Center = j –
2 Xe Xd
Vt2 1 1
Radius = j +
2 Xe Xd
P (pu)
Fig. 19. SSSL with manual operation
Reference [5] introduced a technique that allows plotting of
the SSSL when an AVR or an AVR-PSS combination is
active in the system. Fig. 20 shows the various SSSLs
(manual, AVR, and AVR-PSS) for a system corresponding to
Fig. 17 with the characteristic values in Fig. 17. The following
comments are worth mentioning:
• All three stability limits go through the point
–1/Xd.
• The AVR stability limit expands the manual limit,
provided the AVR gain is limited. Beyond a gain
sin 2δ
threshold, the SSSL with an AVR will infringe into
(24)
the stability area of the manual case [5].
• The AVR-PSS combination allows an increased gain
of the AVR to a higher value (100 in the example).
The SSSL of the AVR-PSS combination goes much
lower in the underexcitation region than the two other
 7

limits (manual and AVR). The SSSL has, therefore, We can simplify the equation further by removing the
been substantially improved. saliency and replacing E′q with E′i so that we get:
3 E i′ E s
P= sin δ (26)
Vt = 1.0 Xd = 1.6 Xq = 1.55 Xd' = 0.32 X ′d + X e
Xe = 0.4 Te = 0.05 H = 3 T'd0 = 6
As an example, consider the case of a generator operated at
Kstab = 5 Tw = 1.45 T1 = 0.4 T2 = 0.033 unit terminal voltage and unit current with a power factor of
2
0.91. Equation (27) shows the generator impedances in pu
values:
Unit Circle
X d = 1.15, X ′d = 0.37, x q = X ′q = 0.75 (27)
Equation (28) shows the external impedance:
1 X e = 0.2 (28)
For this condition of operation, we can compute the
Q (pu)

generator internal and infinite bus voltages in Fig. 18.

Manual SSSL E q = 1.793
0 1 E ′q = 1.179
– (29)
Xd E i′ = 1.345
E s = 0.935
Fig. 21 shows the three power transfer curves: one for the
–1
steady state and corresponding to (24) and two for the
transient state corresponding to (25) and (26). In transient
AVR SSSL state, we can apply the equal-area criterion on the two
AVR-PSS SSSL
Ke = 10
Ke = 100 transient state curves. One can see that the two curves in the
–2
transient state exhibit a much higher peak value, so we can
0 0.5 1 1.5 2 2.5 expect a better transient stability if we were to use the curve in
P (pu) steady state.
Fig. 20. Various SSSLs depending upon the nature of the exciter 2.5

B. Transient Stability Transient State Transient State

Constant E'i Constant E'q
Transient stability is the ability of the power system to
maintain synchronism when subjected to a severe transient 2
disturbance such as a fault on a transmission line, loss of
generation, or loss of a large load [2].
Modern techniques allow us to use such tools as transient
1.5
Power P (pu)

stability programs or electromagnetic transients programs

(EMTPs) to study power network transient stability. These
two program types introduce extensive models of generators.
Classical methods to study transient stability have used the 1
equal-area criterion and have necessitated a power transfer
Steady
equation in the transient state.
State
We can obtain a formula for the power transfer equation in
transient stability by replacing Eq with E′q and Xd with X′d in 0.5
(24):
E ′q E s X ′d − X q
P= sin δ + E s2 sin 2δ (25)
X ′d + X e 2 (X ′d + X e ) (X q + X e ) 0
0 40 80 120 160
Angle δ (degrees)

Fig. 21. Generator power angle curves in steady and transient states
 8

IV. CLASSICAL SWING IMPEDANCE CHARACTERISTIC
A. Classical Model of Generators
The classical swing impedance theory determines the
impedance trajectory in the complex plane when the generator
is represented as a constant voltage source, the angle of which
varies with respect to an infinite bus (Fig. 22). This model of a
generator is the same as the classical model shown in Fig. 9.
In Fig. 22, we assumed that Xtr between the generator and the
infinite bus represents the impedance of the step-up
transformer and that Ze corresponds to the series impedance of
a transmission line.
X′d Vt Xtr Ze
ZT
E′i∠δ Es∠0
However, if the excitation system is an AVR, the excitation
will sustain, and may boost, the generator internal voltage so
that ratio n becomes greater than one. The change of the
generator internal voltage depends upon the AVR
characteristics: a slow-acting AVR could limit the rise of the
internal voltage, but a fast-acting AVR could contribute to a
rapid buildup of the internal voltage [3]. After the internal
voltage builds up and ratio n becomes greater than one, the
trajectory circles will move to the upper part of the impedance
plane, with the circles decreasing as n increases (as shown in
Fig. 24).
X
B
Electrical
Ze Center
δ = 180°
ZT

E′j > Es
Xtr

Fig. 22. Generator classical model E′i > Es

δ P
B. Basic Theory of Swing Impedance R
For the network model corresponding to Fig. 22, the X′d
E′i > Es
classical theory of swing impedance allows us to determine
the swing impedance characteristics in Fig. 23. The swing
impedance characteristic depends primarily on the ratio n of A
the two source magnitudes: Fig. 23. Loss-of-synchronism characteristic in the impedance plane
E′
n= i (30) 2
Es n = 1.25
When the two source magnitudes are equal (n = 1), the 1.5
n = 1.5
swing trajectory is a straight line perpendicular to the total
impedance ZT segment. The trajectory crosses the ZT segment 1
at its middle point when the phase angle δ between the two n=2
sources is 180 degrees. This point is called the swing center. 0.5
When ratio n is greater than one, the swing impedance is a B
X (pu)

circle in the upper part of the impedance plane. When n is

0
smaller than one, the swing impedance trajectory will also be n=1
a circle, but it will be in the lower part of the impedance plane. A
Fig. 24 shows a family of circles with different ratios n –0.5
n = 0.5
corresponding to the generator and network impedances in
that figure. –1
n = 0.66
As ratio n becomes increasingly greater or smaller than
X'd = j 0.2
one, the circles become smaller. –1.5 Xtr = j 0.16
Ze = 0.08 + j 0.24 n = 0.8
C. Divergence From the Basic Theory
–2
After a disturbance, the assumption that the generator is a
–1.5 –1 –0.5 0 0.5 1 1.5
constant voltage source is only valid for a short time, R (pu)
compared to T′ d0. In reality, the type and the characteristics of
the excitation system in the generator determine the generator Fig. 24. Family of loss-of-synchronism impedance characteristics
internal voltage and, therefore, ratio n.
With manual operation or constant voltage excitation, one V. CASE STUDIES USING AN EMTP
might assume that the generator internal voltage will decrease This section demonstrates how a generator responds after a
after a disturbance. As a result, ratio n will become smaller disturbance when the generator excitation system is controlled
than one; therefore, the impedance trajectory will follow in any of the following ways:
circles in the lower part of the plane with radii that could • Manually
eventually become smaller and smaller. • Via an AVR
• Via an AVR plus a PSS (AVR + PSS)
 9

The generator used in the Real Time Digital Simulator
(RTDS®) represents a set of 4 x 555 MVA generators (in the
model, we consider them as a single 2220 MVA generator).
Fig. 25 is a high-level representation of how the generator
excitation system will be controlled in the RTDS.
Man
Gen
TM
A. Examples of Swings With Manual Excitation.
In the first case, the fault is cleared after 85 ms and the line
reclosed after 500 ms. The system remains stable. Fig. 27
shows the voltages and currents as measured by a protection
device in Line 2.
VGT

Man Ef

VGT
AVR
If
AVR = Auto

ω PSS 1

2
0

PSS = Y

Fig. 25. A high-level representation of the generator AVR and PSS

controllers used in the case studies
We use a static excitation model as the AVR because of the
rapid response of this type of excitation system. The model
does not include a governor model because of the slow
response of the governor compared to the exciter (AVR). The
mechanical torque applied to the generator is kept constant
(torque prefault = torque post-fault).
We obtained the power system used in this test from [2],
Chapter 12, and show this system in Fig. 26.
Fig. 27. Voltage and current profiles for a three-phase fault with
autoreclosing after 500 ms
The voltages and currents shown in Fig. 27 are the
secondary voltages and currents; the CT and PT ratios are 400
and 3,000, respectively. By using the voltages and currents
from Fig. 27, we can obtain the positive-sequence impedance
and plot this in the impedance plane. Fig. 28 is a plot of the
positive-sequence impedance after the breaker is closed
(faulted line switched back into service).

Line 1

Gen TGen Z1 = 31.2 ∠ 84.4°

VGT
TM Infinite
Bus
Line 2
Ef 24/345 kV
Srated = 2220 MVA
Z% = 15 Z1 = 31.2 ∠ 84.4°
VRated = 24 kV
SRated = 2220 MVA
H := 3.5 MWs/MVA
Xd = 1.81
Xd′ = 0.3 Td0′ = 8.00
Xd′′ = 0.23 Td0′′ = 0.03
Fig. 26. Representation of the model power system used for these case
Xq = 1.76
studies

We apply a three-phase fault to the system; after 85 ms, the

fault is cleared. Switching the faulted line out of service clears
the fault. The line is switched back into service after a
predetermined dead time. Once the fault is cleared, the power
system experiences a power swing. The aim of these
simulations is to show how different excitation systems will
affect the stability of the generator and that of the power
system after a system disturbance.
Fig. 28. Positive-sequence impedance for the first case study
Using the data from Fig. 27, we can calculate the positive-
sequence impedance magnitude. Fig. 29 is a plot of the
positive-sequence magnitude. This plot shows how the
positive-sequence impedance varies during a stable swing
condition. The plot also shows that the rate of change of the
impedance (dZ/dt) is not constant during a swing condition.

Fig. 29. Positive-sequence impedance magnitude for the manual excitation
case where the generator maintains synchronism
Fig. 30 shows the generator speed, active power, and
reactive power for this case.
Fig. 30. Response of the generator speed, active power, and reactive power
outputs before, during, and after the applied fault
Remember that there are two power lines in this system, so
even with one line out of service the generator is still
connected to the power system and can export power. We
confirm this by examining the current in the second line
(Line 2 in Fig. 31).
Fig. 31. A plot of the current in the adjacent unfaulted line before, during,
and after the fault
10
The current, and subsequently the power, through the
second line increases. The reason for this increase is twofold:
• When the fault is cleared, the system experiences a
power swing because the kinetic energy stored in the
rotor increases the rotor speed.
• The generator is being supplied with its predisturbance
mechanical power of 1,960 MW.
The machine wants to export its stored mechanical energy
as well as the input mechanical energy. The effect of this is
that the current through the unfaulted, in-service line
increases. This affects the generator terminal voltage because
increasing current through the line results in decreasing
terminal voltage. The terminal voltage decreases because of
the increase in the internal voltage drop (jXd′It) of the
generator. Because the excitation voltage is fixed, Ex =
constant, the generator EMF Eq does not increase. As a result,
the terminal voltage Vt decreases, which means that the
machine cannot export its input and stored energy unless the
current increases. For this case of manual excitation, the
generator has enough synchronizing torques available to
remain in synchronism with the system when the faulted line
is returned to service after a dead time of 500 ms.
In the second case of the manual excitation mode, we apply
the same fault; the fault is again cleared after 85 ms, but the
line is reclosed (returned to service) after one second. This
time, however, the system (generator) does not experience a
stable swing but becomes unstable. Fig. 32 is a plot of the
voltages and currents a protective device would see for this
instance.
Fig. 32. The voltage and current profile for the three-phase fault whereby
the faulted line is reclosed after 1 second (unstable swing case)
If we use the data from Fig. 31, calculate the positive-
sequence impedance, and plot this impedance in the
impedance plane, we get the curves we would expect from the
theory. Fig. 33 demonstrates use of this theory.

Fig. 33. A plot of the positive-sequence impedance, in the impedance plane,
derived from the voltage and current signal in Fig. 32
Also notice that in Fig. 32, as the terminal voltage
decreases, jXd′Id increases and the value of n decreases, which
means that the diameter of the swing circle decreases for each
consecutive swing. Note that when Line 1 reclosed (see
Fig. 32), the machine voltage was lower than the system
voltage (infinite bus) or, stated another way, n < 1.0.
Therefore, the power swing is in the lower-left-hand side of
the impedance plane. This agrees with the theory presented.
Fig. 34. An enlarged plot of the positive sequence impedance
Close examination of Fig. 33 reveals an anomaly in the
curves; they are not perfect, concentric circles, but seem to
have a flat spot. Based on the voltage and current graphs of
Fig. 32, it appears that the machine may become stable after
the breaker has closed and the pole has slipped. For
approximately three cycles, both lines carry nominal current
and the terminal voltage seems stable. However, after three
cycles, the voltage collapses and the machine becomes
unstable.
11
Fig. 35 shows the variation in the positive-sequence
impedance for the different swings. After the pole slip, the
system seems to stabilize when the impedance stabilizes. This
agrees with what we observed in Fig. 34.
Fig. 35. A plot of the positive-sequence impedance magnitude for the case
when the generator becomes unstable
Examining the generator speed, active power, and reactive
power, as in Fig. 35, one can see that the generator becomes
unstable (has zero synchronizing torque) before the circuit
breaker closes. This instability occurs because the huge drop
in interval voltage inside the generator leaves the voltage at
the terminal Vt so low that the generator cannot export power.
Therefore, the generator electrical output does not match the
mechanical input power. The imbalance between the
mechanical power and electrical power causes energy to be
stored in the rotor in the form of kinetic energy, resulting in
rotor acceleration. This is one of the major drawbacks of a
manually controlled exciter: it cannot boost excitation voltage
after the fault is cleared. If the generator internal voltage Ei
can increase, thereby reducing the internal voltage drop, the
terminal voltage Vt increases because the internal voltage drop
(jXd′It) decreases as a result of the reduced current.
For this case, the critical reclosing time of Line 1 is
600 ms, or 36 cycles. At this time, the machine synchronizing
torque equals zero. We can see that the machine speed begins
to increase and continues increasing, even when the breaker is
reclosed. In practice, overspeed, overcurrent, or pole-slip
protection would have probably tripped the generator/
machine.
Fig. 36. Plot of the generator speed, active power, and reactive power when
the generator experiences an unstable power swing
 12

Another factor to consider when using manual excitation

on a generator is the reclosing time of critical lines. As shown
in the second case, manual excitation can lead to the generator
becoming unstable after a system disturbance.
B. Examples of Swings With AVR Enabled
For this case study, the AVR is enabled and we subject the
system to the same fault and reclosing time that caused the
manually excited system to become unstable. This study will
help demonstrate the effect of the AVR.
Fig. 37 is a plot of the voltage and current waveforms for
the protected line.

Fig. 38. A plot of the positive-sequence impedance in the impedance plane

for the case where the AVR is enabled

The impedance plot in Fig. 38 is further proof that the

swing is stable. Note, however, that the trajectory of this plot
is not as smooth as that of the manual excitation case in
Fig. 28. Instead, this trajectory has discontinuity in its
derivative because the AVR tries to keep the terminal voltage
within defined limits. When the AVR tries to maintain the
voltage within these limits, it regulates the reactive power the
generator exports or imports. This is reflected in the
Fig. 37. A plot of the voltage and current with the AVR enabled for the case impedance plane in terms of the reactive component (X). This
that caused instability in the manually excited generator anomaly is not visible in the positive-sequence impedance
From the voltage and current plot, we can see that the magnitude plot in Fig. 39, because Fig. 39 reflects the
system experiences a stable power swing after the line is magnitude changes caused by the active power decreasing (R
reclosed. What is different about the plot in Fig. 37 compared is increasing), while the reactive power remains almost
to the plots in Fig. 27 and Fig. 32 is that the voltage in Fig. 37 constant (X remains constant).
does not begin to collapse when the fault is cleared. This has a
very important effect because the terminal voltage stays high,
even though the voltage drop (jXd′It) in the machine is
approximately the same as in the manually excited case. This
means the machine can export the same amount of electrical
power as it receives in mechanical input power. Therefore, no
uncontrolled acceleration occurs. Terminal voltage maintains
its prefault value because the AVR boosts the generator
internal voltage. Using the data from Fig. 37, we can calculate
the positive-sequence impedance after the line comes back
into service. This is the impedance such as a distance relay Fig. 39. A plot of the positive-sequence magnitude for the case where the
protecting the line would see. AVR is enabled
Fig. 38 is a plot of the positive-sequence impedance in the
impedance plane. The plot has a time frame of approximately
100 cycles (1.67 seconds) and begins when the breaker
recloses. After approximately 15 seconds, the positive-
sequence impedance reaches its value, which is identical to its
prefault value, indicated as “Load” in Fig. 38.
 13

Fig. 40 is the response of the generator for the applied
conditions. Initially, the speed, active power, and reactive
power are at a constant level (prefault condition). During the
fault, the machine speed increases. This is to be expected
because the amount of electrical active power the generator
exports is not the same as the amount of mechanical power it
imports. Also, during the fault the active power export
decreases and the reactive power export increases, which
agrees with the theory. Once the fault clears, the machine
speed begins to stabilize and the machine begins to export
more active power (MWs) than it did during the prefault
condition. This is because the machine has stored up kinetic
energy in the rotor during the fault, energy that is now also
being exported to the power system. Because the machine is
exporting more electrical power than it is receiving
mechanical power, the rotor speed begins to decrease as the
generator uses up the stored kinetic energy. Up to this point
(±27 cycles after fault clearance [450 ms]), the machine
behaves almost identically for the manually excited and the
automatically excited case. Moving forward, however,
differences begin to appear. In the manually excited case, the
generator active power export begins to decrease, which
results in increasing speed. This downward spiral continues
until the machine slips a pole and becomes unstable. The
instability occurs because the increased voltage drop inside the
generator collapses the terminal voltage, preventing the
generator from exporting the electrical active power that it is
receiving in the form of mechanical energy. When the AVR is
enabled, it keeps the terminal voltage at a more or less
constant level after the fault is cleared.
Fig. 41 shows the response of the AVR. The three-phase
fault on the line reduces the terminal voltage to near zero
during this fault, and the AVR tries to compensate for the low
terminal voltage by boosting the generator internal voltage
(Ei). During the fault, the AVR reaches the voltage maximum
limit (see Fig. 12).
Fig. 41. Output of the AVR
This boost in internal voltage helps maintain generator
stability during the time the faulted line breaker is open,
allowing the generator to export the same amount of active
power as the amount of mechanical power it imports. As
mentioned before, the current the generator exports for the
first 27 cycles after the fault is cleared is almost similar for
both cases (see Fig. 42), but after this point, the manually
excited generator current continues to increase because the
machine is trying to export real power. So, even though
reactive power is linked to voltage, voltage is necessary to
export active power from a generator. Thus, the boost in
internal voltage resulting from the AVR gives the generator
the synchronizing torque to maintain synchronism with the
system.

Fig. 40. A plot of the generator speed, active power, and reactive power for
the case where the AVR is enabled Fig. 42. A comparison of the line current in the unfaulted lines for the case
with and without the AVR enabled
From Fig. 40, we can see that the generator does not settle
back to its prefault condition rapidly after the fault is cleared.
As mentioned before, the generator achieves its prefault state
after about 15 seconds.

C. Examples of Swings With AVR and PSS Enabled
For this third and final case, both the AVR and the PSS
will be enabled, and we will subject the system to the same
condition as previously mentioned. Fig. 43 is a voltage and
current plot for the applied fault and the reclose.
Fig. 43. A plot of the voltage and current with the AVR and PSS in service
The voltage and current plot shows that the generator
remains stable during the open interval time and when the
breaker recloses. If we now plot the positive-sequence
impedance (see Fig. 43 and Fig. 44), we see that the swing is
more damped and the impedance does not traverse a large
section of the impedance plane, as was the case when only the
AVR was enabled.
Fig. 44. A plot of the positive-sequence impedance in the impedance plane
for the case where the AVR and PSS are enabled
14
If we enlarge Fig. 44, we see that the impedance variation
after each swing decreases more (see Fig. 45) than for the case
with only the AVR. The plot in Fig. 45 occurs over the same
time frame, 3 seconds or 180 cycles, as the one in Fig. 38.
From this plot, we can clearly see the effect of the PSS. The
PSS results in the impedance returning more rapidly to its
prefault state.
Fig. 45. Enlarged plot of Fig. 44 to better show the trajectory of the positive-
sequence impedance after the line is reclosed
If we examine the positive-sequence magnitude plot (see
Fig. 46), we see that the rate of change of impedance |dZ|/dt, is
initially as large (for the first swing) as for the plot without the
PSS. However, we can see that this is no longer the case for
subsequent swings, and these oscillations are damped out
more rapidly than in the plot without the PSS. Therefore, we
can see that the PSS damps out the oscillations.
Fig. 46. A plot of the positive-sequence impedance magnitude for the case
where AVR and PSS are enabled
If we now examine the generator speed, active power, and
reactive power (Fig. 47), we can see that the machine returns
to its prefault value more rapidly with the PSS also enabled
than for the same case with only the AVR enabled.

Fig. 47. A plot of the machine speed, active power, and reactive power
when the AVR and PSS are both enabled
Fig. 48 shows the output of the PSS and the AVR for the
above condition. At fault inception, the machine initially loses
speed and then gains speed. This initial drop in speed results
in the PSS issuing a negative gain in the excitation. After this
initial decrease in speed, the machine speed increases and the
PSS and AVR both drive the AVR to its upper limit. It is
interesting that the PSS causes the AVR to be driven to the
upper and lower limits more often than when only the AVR is
enabled. This occurs because the PSS wants to drive the
change in speed of the machine to zero.
Fig. 48. A plot of the signals that are input and output from the AVR for the
case with the AVR and PSS enabled
Note that the PSS can damp out the change in speed by
regulating the power the machine exports. This is because the
PSS can control the generator terminal voltage, and the
machine power is regulated so that the oscillation of the
machine cancels out (absorbs) the power system oscillations.
15
VI. CONCLUSIONS
1. For short-circuit studies, where only the fault current
magnitude is of concern, we can use a constant voltage
source instead of a complete generator model, reducing
the complexity of the system model.
2. Generators with manual excitation experience a decrease
of the generator internal voltage (Ei) following a system
disturbance, such as a fault, close to the generator
terminals. This decrease in internal voltage significantly
reduces the generator’s synchronizing ability after the
disturbance.
3. An AVR significantly improves generator steady-state
stability, provided that the gain of the AVR is limited and
that the AVR is not operating at its limits before the
disturbance. The AVR boosts the generator internal
voltage during a system disturbance; this boost increases
the generator synchronizing torque, allowing the
generator to return to synchronism after the disturbance.
4. An AVR will help increase the synchronizing torque of a
generator but will reduce the damping torque. Because of
this, the AVR gain will have to be limited in most cases.
5. A PSS increases the damping torque of a generator and
allows increasing the AVR gain without compromising
the generator’s dynamic stability. A PSS therefore
improves the dynamic stability of the generator more than
an AVR alone.
6. In transient stability studies, generators modeled with
constant excitation will constitute the worst-case scenario
with respect to system stability following a disturbance.
AVRs and PSSs in the generators should substantially
improve system stability.
VII. APPENDIX A: NOMENCLATURE
The nomenclature used for generator mathematical model
symbols corresponds to the convention of [1].
Eq: excitation voltage or steady-state internal voltage
E′q: quadrature component of transient internal voltage
E′ex: exciter voltage
Eex: exciter voltage referred to the armature
Id: direct-axis component of the output current
Iq: quadrature-axis component of the output current
Lff: field circuit self-inductance
Rf: field circuit resistance
Mf: mutual inductance between the field and any armature
phase
ra: armature phase resistance
T′d0: direct-axis transient open-circuit time constant
T′d: direct-axis transient short-circuit time constant
Vd: direct-axis component of the terminal voltage
Vq: quadrature-axis component of the terminal voltage
Xd: direct-axis synchronous reactance
Xq: quadrature-axis synchronous reactance
X′d: direct-axis transient reactance
ψf: field circuit flux linkage
ψd: direct-axis component of the air-gap flux
ψq: quadrature-axis component of the air-gap flux
ω: generator speed in rad per second
 16

VIII. APPENDIX B but ω = 1, so

In general, the per unit system is well understood; the eq = M fu • I'p
apparatus MVA rating is used as the base MVA and the
nominal voltage is used as the base voltage. From these, then, For the reciprocal per unit system, we can calculate the
we can calculate the base impedance and currents. field current, I′p, required to generate the rated stator voltage
For a generator, the situation is more complicated in that as follows:
we have a field winding (excitation winding on the rotor) and 1
an armature winding (on the stator). The field winding is I′f := (pu)
excited from an excitation system/circuit. The excitation M fu
system has a per unit value based on the field current required This generates the corresponding field current:
to produce one per unit voltage on the stator along the air-gap
Rf
line. The generator per unit system is based on its apparent E ex ′ = R f • I′p =
power rating (MVA) and its nominal rating (kV) [base2]. M fu
However, we need to match the exciter with the generator, and By definition, we know that corresponding exciter output
this leads to a dilemma: what per unit system do we use when current If is 1.0 pu.
modeling a generator system? Therefore, If := Mfu • I'f with a corresponding exciter
We find that the generator has two per unit systems: voltage output:
• Reciprocal per unit system: Used for modeling the
M fu
generator armature (main machine) and the system, E f := • E ex ′
based on the generator MVA rating and nominal Rf
terminal voltage. Remember that, physically, the exciter output voltage and
• Nonreciprocal per unit system: Used to model the current are the same as the generator-field current and voltage.
generator excitation system. One per unit exciter The exciter is connected via slip rings to the generator’s field
voltage (field voltage) is equal to one per unit winding. We only make the distinction in the per unit system
armature voltage (terminal voltage) along the air-gap to allow the independent selection of a per unit system, thus
line. allowing for modeling of the generator excitation system and
the main generator. Fig. 50 is a simple sketch illustrating how
Air Gap to convert from the reciprocal per unit system to the non-
Vt (pu) reciprocal per unit system and vice versa.
OCC
Physical Unit
1.0
Conversion Generator
Exciter Model R Model
Eex′ Eex′ = f • Eex′
Mfu Eex′
If
I I′f
Slope = Mfu I′f = f
Nonreciprocal Mfu Reciprocal

Fig. 50. Per unit conversion between excitation system and synchronous
machine

However, to refer physical units (actual values) from one

system to the other, we have to use a different approach. We
1
(1.0) I′p , (If)
Mfu know the following:
Ef
Fig. 49. Synchronous machine open circuit characteristic E q := ω • M f • If and If =
Rf
Let us determine a relationship between the current and
voltage in nonreciprocal and reciprocal per unit systems. To Mf M
E q := ω • • E f and let γ = ω f
do this, let us examine an unloaded (open circuited) generator, Rf Rf
where id = iq = 0.
We know the following: We can now relate Ef to Eq via γ, which is similar to
relating the voltage of one winding of a transformer to that of
ed := −ψ q = Lq • i q = 0
(iq = 0) another winding:
eq := ψ d = Ld • i d + M f ⋅ i f = M f • i d (id = 0)  n1 
If we now look at Fig. 49, we see that the field current  E1 = • E2 
 n2 
needed to produce one per unit of terminal voltage (Et) on the
air-gap line is given by:
Vt := eq = ω • M fu • I'p = 1.0 pu
 17

The following sketch illustrates how to move from the non- Now let us revisit some generator flux equations:
reciprocal system to the reciprocal system and vice versa. ψ d = M f • I′f − L d • i d (38a)

Physical Unit ψ q = −Lq • iq (38b)

Conversion Generator
Exciter Model
Model 3
Eex′ M
Eex′ = ω f • Ef Eex′
ψ f = Lff • I′f − • M f • i d (38c)
If Rf 2
I′f
I′f = Mfu • If Let us examine the difference between the quadrature axis
Nonreciprocal Reciprocal
voltages, Eq and Eq:
Mf
Fig. 51. Physical unit conversion between excitation system and E q − E q ' = ω • M f • I′f − ω • •ψf
synchronous machine Lff
A. Transient State Analysis [using (34) and (36)]
Let us now examine the excitation circuit when the (39)
generator is subjected to a system transient such as a power If we substitute (38c) into (39), we get the following:
system fault or a switching operation:
ω • Mf  3 
dψ f E q − E q ' = ω • M f • I′f − •  Lff • I′f − • M f • i d 
E ex ′ = R f • I′f + (nonreciprocal base) (31) Lff  2 
dt
Mf ω • Mf 3 ω • Mf 2
E ex ′ = ω • E ex ′ (reciprocal base) (32) = ω • M f • I′f − • L ff •I′f + • • id
Rf Lff 2 Lff
(40)
M M M dψ 3 ω • Mf 2
ϖ f • E ex ′ = ω f • R f • I′f + ω f • f (33) = • • id
Rf Rf R f dt 2 Lff
[converting (31) to a reciprocal base] If we know that the field is closed but not energized
We know the following: (ψf = 0), and we apply this to (38c), we get the following:
E q = ω • M f • I′f (34) I′f =
3 Mf
• id (41)
2 Lff
L
Tdo ' = ff (35) Substituting (41) into (38a), we get the following:
Rf
3 M
During a fast change of armature current (fast compared to ψ d = M f • • f • id − Ld • i d
transient decrement, but not faster than the subtransient 2 L ff
decrement), the flux linkage, ψfd, of the field remains
ψd 3 M2
substantially constant (i.e., dψfd/dt ≈ 0). This nearly constant = Ld − • f
flux linkage results in a new fictitious internal armature −i d 2 L ff
voltage proportional to the field flux linkage: = Ld ' (for a generator, for a motor, ψ d = Ld • i d )
M
Eq ' = ω • f • ψf (36) 3 Mf 2
Lff • = Ld − Ld ' (42)
2 L ff
When we substitute (35) and (36) into Part C of (33), we
Substituting (42) into (40), we get the following:
get the following:
Mf • ψf L M
E q − E q ' = ω • ( Ld − Ld ') • id
ω• = ff • ω • f • ψ f = Tdo′ • E q′
Rf Rf Lff E q − E q ' = ( x d − x d ') • id (43)
Substituting (32) into Part A and (34) into Part B of (33),
we obtain the following:
Eq '
E ex ′ = E q + Tdo ' •
dt
Eq ' E ex ′ − E q
= (37)
dt Tdo '
 18

If we now wrap all of this into a phasor diagram, we get the [10] ANSI Standard C50.13-1989, Requirements for Cylindrical Rotor
following: Synchronous Generators, 1989.
[11] ANSI Standard C50.12-1982, Standard for Requirements for Salient-
(xd – xd′)Id Pole Synchronous Generators and Generator/Motors for Hydraulic
Turbine Applications, 1982.
E
Ed [12] R. Sandoval, A. Guzmán, and H. J. Altuve, “Dynamic Simulations Help
Quadrature Improve Generator Protection,” proceedings of the 33rd Annual Western
Iq Eq′ Eq Axis
Protective Relay Conference, Spokane, WA, October 2006.
δ
Ed′ E′ xqIq
xd′Id X. BIOGRAPHY
Φ xq′Iq Ei′
Vt xd′I Normann Fischer received a Higher Diploma in Technology, with honors,
rI xdId from Witwatersrand Technikon, Johannesburg in 1988, a B.S.E.E., with
honors, from the University of Cape Town in 1993, and an M.S.E.E. from the
Id I University of Idaho in 2005. He joined Eskom as a protection technician in
1984 and was a senior design engineer in Eskom’s Protection Design
Direct Department for three years. He then joined IST Energy as a senior design
Axis engineer in 1996. In 1999, he joined Schweitzer Engineering Laboratories,
Inc. as a power engineer in the Research and Development Division. He was a
Fig. 52. Phasor diagram of a synchronous machine during the transient state registered professional engineer in South Africa and a member of the South
Africa Institute of Electrical Engineers.
IX. REFERENCES
[1] E. W. Kimbark, Power System Stability: Synchronous Machines, Dover Gabriel Benmouyal, P.E. received his B.A.S. in Electrical Engineering and
his M.A.S. in Control Engineering from Ecole Polytechnique, Université de
Publications, Inc., New York.
Montréal, Canada, in 1968 and 1970, respectively. In 1969, he joined Hydro-
[2] P. Kundur, Power System Stability and Control, New York: McGraw- Québec as an Instrumentation and Control Specialist. He worked on different
Hill, 1994. projects in the field of substation control systems and dispatching centers. In
[3] G. Benmouyal, “The Impact of Synchronous Generators Excitation 1978, he joined IREQ, where his main field of activity was the application of
Supply on Protection and Relays,” proceedings of the 34th Annual microprocessors and digital techniques for substation and generating-station
Western Protective Relay Conference, Spokane, WA, October 2007. control and protection systems. In 1997, he joined Schweitzer Engineering
Laboratories, Inc. in the position of Principal Research Engineer. He is a
[4] IEEE Standard 421.5-1992, IEEE Recommended Practice for Excitation
registered professional engineer in the Province of Québec, an IEEE Senior
System Models for Power System Stability Studies, 1992.
Member, and he has served on the Power System Relaying Committee since
[5] F. P. DeMello and C. Concordia, “Concepts of Synchronous Machine May 1989. He holds over six patents and is the author or coauthor of several
Stability as Affected by Excitation Control,” IEEE Transactions on papers in the field of signal processing and power networks protection and
Energy Conversion, Vol. PAS−88, No. 4, pp. 316−329, April 1969. control.
[6] IEEE Task Force on Excitation Limiters, “Underexcitation Limiter
Model for Power System Stability Studies,” IEEE Transactions on Satish Samineni received his B.E. degree in electrical and electronics
Energy Conversion, Vol. 10, No. 3, September 1995. engineering from Andhra University College of Engineering, Visakhapatnam,
[7] D. Reimert, Protective Relaying for Power Generation Systems, Boca India. He received his Master’s degree in Electrical Engineering from the
Raton: CRC Press, 2006. University of Idaho in 2003. Since 2003, he has been with Schweitzer
Engineering Laboratories, Inc. in Pullman, Washington, where he presently is
[8] IEEE Power Engineering Society, IEEE Tutorial on the Protection of a Lead Power Engineer. His research interests include power electronics and
Synchronous Generators, 95 TP 102. drives, power system protection, synchrophasor-based control applications,
[9] IEEE Standard C37.102/D7−200X, Guide for AC Generator Protection, and power system stability.
April 2006.

© 2008, 2011 by Schweitzer Engineering Laboratories, Inc.

All rights reserved.
20111220 • TP6328-01